MuSR
MuSR (Multistep Soft Reasoning) is a benchmark for evaluating language models on multistep soft reasoning tasks specified in natural language narratives. Created through a neurosymbolic synthetic-to-natural generation algorithm, it generates complex reasoning scenarios like murder mysteries roughly 1000 words in length that challenge current LLMs including GPT-4. The benchmark tests chain-of-thought reasoning capabilities across domains involving commonsense reasoning about physical and social situations.
Kimi K2 Instruct from Moonshot AI currently leads the MuSR leaderboard with a score of 0.764 across 2 evaluated AI models.
What MuSR measures
MuSR is a text benchmark that evaluates large language models on reasoning tasks. LLM Stats tracks 2 models on this benchmark, with a maximum possible score of 1. Current average across reported models is 0.6, with the leader reaching 0.8.
Compare leaders on the best AI for reasoning leaderboards.
Publication
- Paper
- MuSR: Testing the Limits of Chain-of-thought with Multistep Soft Reasoning
- Authors
- Zayne Sprague, Xi Ye, Kaj Bostrom, Swarat Chaudhuri, and 1 others
- Published
- arXiv
- 2310.16049
Abstract
While large language models (LLMs) equipped with techniques like chain-of-thought prompting have demonstrated impressive capabilities, they still fall short in their ability to reason robustly in complex settings. However, evaluating LLM reasoning is challenging because system capabilities continue to grow while benchmark datasets for tasks like logical deduction have remained static. We introduce MuSR, a dataset for evaluating language models on multistep soft reasoning tasks specified in a natural language narrative. This dataset has two crucial features. First, it is created through a novel neurosymbolic synthetic-to-natural generation algorithm, enabling the construction of complex reasoning instances that challenge GPT-4 (e.g., murder mysteries roughly 1000 words in length) and which can be scaled further as more capable LLMs are released. Second, our dataset instances are free text narratives corresponding to real-world domains of reasoning; this makes it simultaneously much more challenging than other synthetically-crafted benchmarks while remaining realistic and tractable for human annotators to solve with high accuracy. We evaluate a range of LLMs and prompting techniques on this dataset and characterize the gaps that remain for techniques like chain-of-thought to perform robust reasoning.
Kimi K2 Instruct leads with 76.4%, followed by 
Hermes 3 70B at 50.7%.
Progress Over Time
Interactive timeline showing model performance evolution on MuSR
MuSR Leaderboard
| Context | Cost | License | ||||
|---|---|---|---|---|---|---|
| 1 | Moonshot AI | 1.0T | — | — | ||
| 2 | Nous Research | 70B | — | — |
FAQ
Common questions about MuSR.
More evaluations to explore
Related benchmarks in the same category
A challenging dataset of 448 multiple-choice questions written by domain experts in biology, physics, and chemistry. Questions are Google-proof and extremely difficult, with PhD experts reaching 65% accuracy.
A more robust and challenging multi-task language understanding benchmark that extends MMLU by expanding multiple-choice options from 4 to 10, eliminating trivial questions, and focusing on reasoning-intensive tasks. Features over 12,000 curated questions across 14 domains and causes a 16-33% accuracy drop compared to original MMLU.
All 30 problems from the 2025 American Invitational Mathematics Examination (AIME I and AIME II), testing olympiad-level mathematical reasoning with integer answers from 000-999. Used as an AI benchmark to evaluate large language models' ability to solve complex mathematical problems requiring multi-step logical deductions and structured symbolic reasoning.
Massive Multitask Language Understanding benchmark testing knowledge across 57 diverse subjects including STEM, humanities, social sciences, and professional domains
A verified subset of 500 software engineering problems from real GitHub issues, validated by human annotators for evaluating language models' ability to resolve real-world coding issues by generating patches for Python codebases.
Humanity's Last Exam (HLE) is a multi-modal academic benchmark with 2,500 questions across mathematics, humanities, and natural sciences, designed to test LLM capabilities at the frontier of human knowledge with unambiguous, verifiable solutions